This project on the epidemiology of alcohol-related problems will integrate data from multiple sources to investigate environmental correlates of underage drinking and adult problem drinking. The relationship between alcohol outcomes and neighborhood environments has been difficult to determine using individual- level population data because human subjects constraints restrict the release and use of geographic identifiers. If location codes are available at all in larger data sets, they may be as crude as a state or metropolitan area identifier. The proposed analytic tools and data are also new in this field. Our objectives coincide with the objectives of Secondary Analysis of Existing Alcohol Epidemiology Data Sets (PA-06-557) in two dimensions: (a) analyzing previously collected data to advance scientific knowledge of alcohol-related epidemiology and (b) applying new approaches to analyze existing data. Our specific aims are to: 1. Integrate individual-level data from the 2001, 2003, and 2005 waves of the California Health Interview Survey (CHIS) with data on the respondents' environment. We will derive environmental characteristics for each respondent for various geographic units, beginning with areas smaller than census tracts. 2. Study the relationship between neighborhood characteristics and a) underage drinking and b) problem drinking among adults. We are particularly interested in estimating what size geographic unit is most predictive of alcohol misuse. 3. Document disparities in environmental characteristics that may affect and interact with alcohol outcomes across sociodemographic groups. 4. Examine the extent to which contextual protective factors such as collective efficacy and school environments moderate problem drinking. [unreadable] [unreadable] Public Health Relevance: The project analyzes four important questions for alcohol research and policy: Which environmental and neighborhood characteristics predict problem drinking? How narrow or large is the relevant neighborhood that affects alcohol outcomes? How does that differ for adults versus adolescents? Do these environmental characteristics predominantly disadvantage certain population subgroups? [unreadable] [unreadable] [unreadable]